Language and Ethnic Statistics in 20Th Century Sudanese Censuses and Surveys Catherine Miller
Total Page:16
File Type:pdf, Size:1020Kb
Language and ethnic statistics in 20th century Sudanese censuses and surveys Catherine Miller To cite this version: Catherine Miller. Language and ethnic statistics in 20th century Sudanese censuses and surveys. 2016. halshs-01544597 HAL Id: halshs-01544597 https://halshs.archives-ouvertes.fr/halshs-01544597 Preprint submitted on 23 Jun 2017 HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. Language and ethnic statistics in 20th century Sudanese censuses and surveys Catherine Miller, Aix Marseille University Abstract The paper investigates the creation of language statistics in the Sudan, from the beginning of the 20th century up to the division of the country into two states. Like many other African countries, Sudan is characterized by a high degree of ethnic and linguistic diversity that has participated in the fueling of murderous civil wars since independence. The paper recontextualizes the construction of the ethno-linguistic categories and statistics within their broader political and administrative contexts. It analyzes the objectives and output of each type of statistics and questions their influence on the foreign and native representations of Sudanese society. Key words: Sudan, ethno-linguistic categories, statistics, national census, colonial and post linguistics. 1 Introduction Within African linguistics circles, Sudan is known for its linguistic diversity, not only in terms of numbers of languages but also because it contains languages belonging to three of the four language families attested in Africa per Greenberg’s classification (i.e. Afro-Asiatic, Nilo-Saharan, and Niger-Kordofanian, see Greenberg 1963). In the first part of the 20th century, Sudan therefore occupied a key position within African language classification. The current prevailing Sudanese ethno-linguistic categories have been regarded as established scientific fact (Abdelhay 2010; Abdelhay, Makoni, and Makoni 2010; Miller 2015). It is only recently that, following constructionist and post-colonial studies such as Errington (1998), Gal and Irvine (1995) etc., a few authors have started to describe the ideological implications of the language-making processes of the colonial enterprise, particularly regarding the concepts of ‘indigenous’ or ‘local’ languages (Abdelhay 2008, Abdelhay 2010; Abdelhay, Makoni, and Makoni 2010 and Abdelhay, Makoni, and Makoni 2016). Ethno-linguistic categories represent major contemporary political and social issues in Sudan due to the intensity of the conflicts that have plagued the country since independence and led to the division of the country into two states in 2011. Language policies have been among the factors fueling the long Sudanese civil wars (Abdelhay 2008; Abu Bakr 1995; Hurreiz 1989; Lesch 1998; Miller 1986, Miller2003; Nyombe 1994; Rondyang 2007; Sharkey 2003 and Sharkey 2008). The two main peace agreements between the South and the North (Addis Ababa in 1972 and Naivasha in 2005) included important language decrees concerning the status of the different Sudanese languages (Abdelhay 2008; Abdelhay, Makoni, and Makoni 2011; Berair 2007). It is in front of this tense political background that the making of language statistics in Sudan needs to be examined. Censuses and statistics are key tools in the construction of social categories (Anderson 1991, Appadurai 1993). As Kertzer and Arel pointed out: “Rather than view social links as complex and social grouping situational, the view promoted by the census is one in which populations are divided into neat categories” (Kertzer and Arel 2002: 6). This paper presents a short chronological overview of the main language and ethnic statistics in Sudan through a variety of sources: the 1928 Rejaf Language Conference, the Sudanese national censuses, the language surveys undertaken by the Institute of African and Asian Studies (IAAS) in Khartoum, and the statistics provided by the Summer Institute of Linguistics (SIL) via its database Ethnologue. Each type of statistics reflects different historical contexts, as well as different logics of classification, different institutional or political goals. This has led to rather large discrepancies between statistics in terms of the number of Sudanese languages and the number of speakers recorded. According to the main references, the number of recorded Sudanese languages is 106 (Tucker and Bryan 1956), 135 (Mugaddam and Dimmendaal 2006), 136 (Greenberg 1963; Bell 1975), 142 (Ethnologue), or 177 (Abu Bakr and Hurreiz 1984). The variation in the number of languages reflects a tension between two main trends of language categorization and statistics: reproducing large language groups or language clusters versus reproducing small language units. We find here what Irvine (2008: 338) has characterized as “a process of lumping together or splitting apart linguistic varieties in order to produce territorially regimentable language boundaries.” Whenever possible, the paper tries to identify the sources, objectives and ideology of each type of statistics. Generally speaking, large language groups have been established for educational and administrative purposes whereas smaller language units serve as a tool of identification and belonging. The use of language and ethnic statistics is also closely associated with administrative and political decisions by government, particularly with what has been known in Sudan as the Native Administration system, established during the colonial period and reactivated in 1994 by the Sudanese Islamist government. Since the early 20th century, the conducting of language statistics in the Sudan has been marked by two important ideological stances. First, many statistics and classifications postulate a close and natural association between languages, ethnic groupings, and territorial belonging. This postulate is part of the heritage of the 19th century European scientific background (Gal and Irvine 1995), which has dominated linguistics all around the world. Second, many statistics tend to reproduce the view that in Sudan, there is a specific linguistic division between Arabic - typically treated as one single language - and all other Sudanese languages, labeled African, local, vernacular or indigenous languages. 2 Speakers or Tax payers? Language statistics in South Sudan during the Colonial era 2.1 Native Administration and the North-South Policy Ethno-linguistic classifications and statistics played a crucial role as tools of governance during the colonial era (1898-1956), particularly in South Sudan. In 1921, the British colonial administrative policy implemented the Indirect Rule and Native Administration (NA) system in both North and South Sudan (Abu Shouk 1998). The ‘native’ population was to be represented along tribal lines, with the idea that the bigger, older, and more powerful groups had the right to have appointed representatives in the various levels of administration. In each region, the major tribal groups were also supposed to represent the smaller groups that were administratively affiliated to them. This administrative construction determined access to land rights and natural resources. The first task of the colonial power was therefore to identify the tribes, groups, and sub groups. The colonial archives abound with administrative reports trying to assess the different ethnic/tribal groups of each area and the demographic weight of the tribes. In Northern Sudan (including Western and Eastern Sudan), the colonial classification often reproduced previous tribal-administrative classifications established by the former Sudanese kingdoms like the Funj (Spaulding 1985) or by the Ottoman-Egyptian administration of the 19th century. For Arab groups, the classification was based on tribal genealogy (MacMichael 1922). For non-Arab groups, the classification was based on a number of factors such as genealogy, physical anthropological features, and languages. In South Sudan, the Native Administration system was immediately influenced by an anti-Arab and anti-Islam policy that determined a specific educational and language policy. Language classification and ethno-linguistic statistics became important administrative tools of this policy. Identification of ethno-linguistic groups was undertaken by British district officers, anthropologists such as Evans Pritchard, linguists such as D. Westermann and A.N. Tucker, and missionaries from the different congregations that had been given religious and educational control of assigned territorial areas. From the very beginning, the British Condominium had established different systems of education in North and South Sudan. In the dominant Muslim North, the British were careful to maintain Arabic as language of instruction in primary schools to avoid the upheaval of Mahdist or other Islamist movements. A very different policy was implemented in the South (including the Nuba Mountains in South Kordofan, Abdelhay 2010). Education was left in the hand of the missionaries who opened village schools using local vernacular languages as media of instruction (Beshir 1968